Evaluation of MCP Correlation Algorithms Applied to Wind Data Series
A. Moreira(1), T. Rocha(2), J. Mendonça(3), R. Pilão(4) and P. Pinto(1) 1. MEGAJOULE S.A., Maia (Portugal) 2. INESC TEC, Centre for Power and Energy Systems, Porto (Portugal) 3. SIIS, Instituto Politécnico do Porto, Porto (Portugal) 4. CIETI, Instituto Superior de Engenharia do Porto (Portugal) 2026-01-20

Abstract

This work aimed to develop methodologies foranalysing statistical correlations among wind data series using various Measure-Correlate-Predict (MCP) methods, with the goal of selecting the most suitable method for extrapolating long-term data with minimal associated uncertainty. It was analysed the minimum time required for a wind measurement campaign when applying this methodology. Fifteen local wind measurement stations were selected. The long-term wind data reanalysis series that exhibited the strongest correlation with the measured wind data at each station was then chosen. Multiple tests were conducted with different simultaneous periods between the measured data series and the long-term series. Fifteen correlation algorithms were tested for each concurrent period. The performance of each model was evaluated using the RMSE (Root Mean Square Error) and MBE (Mean Bias Error) associated with each MCP. Analysis of the errors identified measurement periods with the lowest associated error ranging from 1 to 5 years and a single-factor ANOVA analysis was conducted. Finally, t-significance tests were performed. The study concluded that the Neural Network was the most effective MCP method. Additionally, it was determined that the minimum number of years required for a local measurement campaign should be between 2 and 3 years.

Key words: MCP, correlation algorithms, local measurement station, wind regime, reanalysis data series

References

[1] J. V. P. Miguel, “Influência da duração da campanha de medição anemométrica na avaliação de recursos eólicos com base na aplicação de métodos de MCP,” Instituto da energia e do ambiente, São Paulo (2016), pp 21-26.

[2] J. T. Jervell, “Estudo da influência das características do vento no desempenho de aerogeradores”, Tese de Mestrado em Engenharia Mecânica, Porto (2008). Porto, pp 8-18.

[3] R. M. T. Gomes, “Análise de Incerteza em Estimativa de Produção de um Parque eólico”, Tese de Mestrado em Engenharia Eletrotécnica e de Computadores, Porto (2018). pp 5-15.

[4] H. L. S. d. Reis, “Estudo sobre a aplicação do método MCP em dados de longo período para a estimativa de recurso eólico e solar”, Brasília (2013). pp 5-20.

[5] L. Landberg, L. Myllerup, O. Rathmann,, E. L. Petersen, B. H. Jørgensen, J. Badger, N. G. Mortensen, “Wind resource estimation—an overview”, Wind Energy: An International Journal for Progress and Applications in Wind Power Conversion Technology (2003). Vol. 6 (3), pp. 261-271.

[6] J. C. Woods, S. J. Watson, (1997). “A new matrix method of predicting long-term wind roses with MCP”, Journal of Wind Engineering and Industrial Aerodynamics (1997). Vol. 66(2), pp. 85-94.

[7] S. A. Kalogirou, “Artificial neural networks in renewable energy systems applications: a review”,. Renewable and sustainable energy reviews (2001). Vol. 5(4), pp. 373-401.

[8] J. Hair, W. Black, B. Babin, R. Anderson, (2019). “Multivariate Data Analysis”, Prentice Hall, (2019).